INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ayne

-0.69

ãĤ¦ãĤ¹

-0.68

ilts

-0.68

Dur

-0.65

cair

-0.65

»Ĵ

-0.63

aida

-0.62

LOD

-0.62

":"/

-0.62

Bir

-0.62

POSITIVE LOGITS

ktop

0.69

pes

0.65

 Amid

0.64

 Sloven

0.63

awan

0.62

hetti

0.60

ardless

0.60

ALLY

0.60

 æľ

0.59

 Granted

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.